Transvaginal ultrasonography to predict preterm birth in women with bicornuate or didelphus uterus.

OBJECTIVE

To estimate whether cervical length measured by transvaginal ultrasonography (TVUS) in women with uterine anomalies predicts spontaneous preterm birth (SPTB).

METHODS

This retrospective cohort study compared women with a uterine anomaly who were pregnant with singleton gestations and delivered August 2000 to April 2008 to a low risk control group. Transvaginal ultrasonographic cervical lengths were measured 16-30 weeks gestation. Primary outcome was cervical length and SPTB less than 35 weeks and the primary exposure variable of interest was cervical length. Secondary outcomes were SPTB less than 37 weeks, less than 32 weeks, low birth weight, maternal and neonatal outcomes. Receiver operating characteristic curves were generated to identify the best cervical length cutoff.

RESULTS

Women with a bicornuate uterus (N = 35) had shorter cervical length (3.46 cm) than the low risk control group (N = 122, 4.32 cm, p < 0.0001). Women with a bicornuate or didelphus uterus, compared with low risk women, had higher rates of SPTB less than 35 weeks (8.6% and 30.8% versus 0.8%, p = 0.0007), neonatal intensive care unit admission more than 24 h (26.5% and 41.7% versus 7.5%, p = 0.0021) and composite perinatal morbidity (32.4% and 69.2% versus 8.3%, p < 0.0001). Using a cutoff of 3.0 cm, TVUS cervical length in women with a bicornuate uterus predicted SPTB less than 35 weeks (positive predictive value [PPV] = 37.5% and negative predictive value [NPV] = 100%), birth weight less than 2500 g (PPV = 50.0% and NPV = 96.3%) and respiratory distress syndrome (PPV = 37.5% and NPV = 100%). CONCLUSION Women with a bicornuate uterus have shorter cervical lengths than low risk controls, and are at higher risk of SPTB less than 35 weeks. Transvaginal ultrasonographic cervical length predicts SPTB less than 35 weeks, low birth weight and perinatal morbidity in these women.